1. Field of the Invention
The present invention relates generally to corona reaction systems and more specifically to corona generators which utilize the reactant gas stream for removing substantially all of the waste heat from the corona discharge gas flow path.
2. Description of the Prior Art
It is generally known that corona generators may be used to induce a variety of chemical reactions. In particular, corona generators have long been used to produce ozone from oxygen. The quantity of ozone or other product produced in a given generator will depend upon many parameters such as oxygen or reactant gas concentration, electric power applied, temperature, gas flow rate, and so forth.
It has been recently proposed that by maximizing the gas flow rate through a corona generator, the waste heat may be rejected in the reactant gas stream while maintaining the temperature of the reactant gas at a tolerable level. In this manner the need is eliminated for a heat exchanger containing a second head absorbing fluid, coextensive and contiguous with the corona and electrodes. Such heat exchangers are bulky and expensive and greatly complicate the design and construction of the corona generator. In the production of ozone in particular, it is extremely critical that the reactant gas temperature be maintained at a relatively low level so as to inhibit the decomposition of ozone and to maintain a reasonable dielectric life. Accordingly, if a large flow of reactant gas is utilized, it is found that the reaction temperature within the corona generator device may be maintained at a temperature below which tolerable amounts of ozone reversibly react to form oxygen. However, it has recently been found that when relatively weak dielectrics such as glass are employed, the volumes of gas needed to avoid temperature destructive to the dielectric are very large and the long restrictive gas passages conventionally employed create or contribute to an undue pressure drop through the device.
An example of corona generators of the above type wherein substantially all of the heat generated by corona discharge is removed from the generator by gas flow therethrough is described in Canadian Pat. No. 689,301 to Ewing et al. This reference describes an ozone generator employing flat plate electrodes 24 inches by 24 inches square with a glass dielectric having a dielectric strength of about 500 volts/mil based on a maximum permissible operating temperature of about 200.degree. F. Ewing et al discloses a specific example using a power density of 20 watts/ft.sup.2, a gap height of 2 mm, and a final product of 1.5 wt.% O.sub.3 obtained in six generator stages.
Operation in the manner taught by Ewing et al has certain disadvantages. First, the pressure drop is so low that uniform flow distribution through the corona discharge gap is not obtained. The gas tends to channel through certain areas of the gap, and becomes stagnant in others. Ozone production is substantially less than that which would be obtained in the absence of such aberrant flow phenomena. Second, the ozone production per unit electrode area, whether actual or predicted on the basis of theoretical considerations, is very low and well below the yield per unit area obtained with the well-known water-cooled, glass tube electrode corona generators. It should be noted that Ewing et al cannot increase the oxygen flow rate without sustaining a corresponding reduction in ozone concentration. For example, if the flow were increased to 100 lbs O.sub.2 /hr, the concentration gradient across the corona discharge zone would decrease to approximately 0.02%, and about 75 generator stages would be needed in order to reach an ozone concentration of 1.5% in the final product. Whereas this is possible theoretically, it is impractical for commercial use, and as a result, the teachings of Ewing et al have not been usefully applied.
Ewing et al also suggests that power density can be increased to 200 watts/ft.sup.2. At such power density values, and based on the same concentration gradient values across the ozone generator as mentioned above, flow, pressure drop and ozone production are correspondingly increased. Thus, the higher power density improves ozone production considerably over the 20 watts/ft.sup.2 case, but the yield is still not significantly better than achievable with water-cooled, glass tube electrode systems. Moreover, the application of 200 watts/ft.sup.2 in reality exceeds the practical operating limit for glass and similar materials, whose dielectric strength is on the order of 500 volts/mil. With such dielectrics, experience has shown that even at 90 watts/ft.sup.2 the failure rate of glass dielectrics is prohibitively high for commercial practice. It is for this reason, low dielectric strength, that the water-cooled, glass tube systems normally operate in the range 20-90 watts/ft.sup.2 in order to achieve a balance between reliability, reasonable ozone yields and low dielectric failure rates.
One of ordinary skill might conclude that capability of the system disclosed in the Ewing et al patent for efficient operation at high power density levels might be marginally improved by employing recently developed porcelain enamel dielectric materials such as are disclosed for example in U.S. Pat. No. 3,891,561 issued Sept. 7, 1973 to F. E. Lowther. According to the Lowther patent, the specific dielectric materials therein disclosed permit operation at power densities of 1.8 watts/in.sup.2 or about 260 watts/ft.sup.2, compared to the 200 watts/ft.sup.2 taught by Ewing et al. Although such combination, of the Lowther dielectric in the Ewing et al corona generator, permits operation at reasonable gas flow rate and pressure drop conditions, the ozone yield efficiency for this system is still little better than that achieved by the prior art glass tube ozone generators. Nonetheless, it has recently been found that the dielectric material disclosed in the Lowther patent has a substantially higher power density capability than the 260 watt/ft.sup.2 power density level disclosed in the patent. Experience has shown that power densities of 1000 watts/ft.sup.2 and higher can be transmitted across such dielectric for extended periods of operation without failure. Unfortunately, however, this capability cannot be realized in ozone generators of a type as disclosed in the Ewing et al patent without encountering prohibitively high pressure drop across the generator.
Another type of prior art ozone generator which does not employ discrete heat exchangers contiguous to the electrode or gap is the small odor-control corona discharge device conventionally used for space conditioning. Typically such devices consist of paired screen electrodes each on the order of 1 inch square, each forming a discharge gap with a mica dielectric interposed between the screen electrodes. Air is blown at a very high velocity through the gap and across the exposed electrode surfaces--the flow rate being sufficient to produce less than 1 ppm (&lt;0.0001%) ozone in the air discharge. Such concentration levels cannot be exceeded because of toxicity considerations.
The above odor-control devices are not concerned with high ozone yield or with high efficiency. Corona power is extremely low, and any dielectric capable of sustaining low minimal sparking voltages will suffice. Heat generation and high temperatures are not a problem regardless of electrode size because the gaps formed between the dielectric and the adjacent sides of each electrode are fully exposed to ambient air. Thus, the dielectric is effectively and adequately cooled by convection and radiation from the electrode surface.
Accordingly, it is an object of the present invention to provide an improved corona reaction system in which heat generated by corona discharge in the reaction system is removed by the reactant gas stream.
It is a further object to provide an ozone generator which is capable of processing large quantities of oxygen-containing feed gas with minimum pressure drop across the generator.
It is still another object to provide an ozone generator which is capable of efficiently producing large yields of ozone.
Other objects and advantages of the invention will be apparent from the ensuing disclosure and appended claims.